sign in sign up
<<
Home , Beta cell, Wnt signaling pathway

Lee et al. Molecular Brain (2016) 9:24 DOI 10.1186/s13041-016-0207-5

RESEARCH Open Access Wnt3a upregulates brain-derived by increasing NeuroD1 via Wnt/β- signaling in the Jaemeun Lee1, Kyungchan Kim1, Seong-Woon Yu1 and Eun-Kyoung Kim1,2*

Abstract Background: Insulin plays diverse roles in the brain. Although insulin produced by pancreatic β-cells that crosses the blood–brain barrier is a major source of brain insulin, recent studies suggest that insulin is also produced locally within the brain. However, the mechanisms underlying the production of brain-derived insulin (BDI) are not yet known. Results: Here, we examined the effect of Wnt3a on BDI production in a hypothalamic line and hypothalamic tissue. In N39 hypothalamic cells, Wnt3a treatment significantly increased the expression of the Ins2 , which encodes the insulin isoform predominant in the mouse brain, by activating Wnt/β-catenin signaling. The concentration of insulin was higher in culture medium of Wnt3a-treated cells than in that of untreated cells. Interestingly, neurogenic differentiation 1 (NeuroD1), a target of Wnt/β-catenin signaling and one of factors for insulin, was also induced by Wnt3a treatment in a time- and dose-dependent manner. In addition, the treatment of BIO, a GSK3 inhibitor, also increased the expression of Ins2 and NeuroD1. Knockdown of NeuroD1 by lentiviral shRNAs reduced the basal expression of Ins2 and suppressed Wnt3a-induced Ins2 expression. To confirm the Wnt3a-induced increase in Ins2 expression in vivo, Wnt3a was injected into the hypothalamus of mice. Wnt3a increased the expression of NeuroD1 and Ins2 in the hypothalamus in a manner similar to that observed in vitro. Conclusion: Taken together, these results suggest that BDI production is regulated by the Wnt/β-catenin/NeuroD1 pathway in the hypothalamus. Our findings will help to unravel the regulation of BDI production in the hypothalamus. Keywords: Brain-derived insulin, Hypothalamus, Wnt/β-catenin signaling, NeuroD1

Background Insulin produced in pancreatic β-cells is a major source Recognition of the importance of insulin action in the of plasma insulin. Insulin produced by the is brain has grown in many aspects such as energy balance, supplied to the brain from the blood circulation through , neuronal survival, synapse forma- the blood–brain barrier [11, 12], since insulin can be tion, and cognition [1–4]. In particular, studies on the transported across this barrier [13]. However, not all insu- relationship between metabolic and neurodegenerative lin in the brain comes from the pancreas, because insulin diseases emphasized the role of insulin signaling in the can also be produced in the brain [14, 15]. Insulin mRNA brain [5–10]. However, the source of insulin in the brain is detected in various regions of the brain both during has not been fully investigated. development and in adults. Rodents have two non-allelic insulin , Ins1 and Ins2 [14, 15]. In the pancreas, both genes are expressed, with Ins2 expressed at a higher level [16]. Interestingly, Ins1 mRNA is not detected in the * Correspondence: [email protected] 1Department of Brain and Cognitive Sciences, Daegu Gyeongbuk Institute of brain. Instead, Ins2 mRNA is found in limbic and olfactory Science and Technology (DGIST), 333 Techno Jungang-daero, regions, hippocampus, and hypothalamus [6, 17–22]. Hyeonpung-myeon, Dalseong-gun, Daegu 42988, South Korea Nevertheless, it is difficult to detect Ins2 mRNA in the 2Neurometabolomics Research Center, Daegu Gyeongbuk Institute of Science and Technology (DGIST), 333 Techno Jungang-daero, brain because of its quite low expression level. Using Hyeonpung-myeon, Dalseong-gun, Daegu 42988, South Korea single-cell digital PCR, one research group found that the

© 2016 Lee et al. Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Lee et al. Molecular Brain (2016) 9:24 Page 2 of 12

neurogliaform cells in the cerebral cortex express Ins2 underlying the regulation of BDI production in the mRNA [23]. To verify the production of insulin in the hypothalamus. brain, brain was stained with antibodies against connect- ing peptide (c-peptide), byproduct of process- Results ing. C-peptide–positive signal was found in the central Wnt3a increases the levels of Ins2 mRNA and in nervous system, and the pattern of c-peptide staining N39 cells showed correlation with that of Ins2 expression [24, 25]. We used the immortalized mouse hypothalamic neuronal However, the mechanisms of local production of insulin cell line N39 (mHypoE-39) to investigate the mechanism in the brain, which is called brain-derived insulin (BDI), of BDI production, because N39 cells express Ins2 [14]. and its roles remain unclear. Consistent with the previous findings [14], Ins2 mRNA Wnt signaling is involved in brain development [26–28]. but not Ins1 mRNA was detected by quantitative real- Wnt are a diverse family of secreted glycopro- time PCR (qRT-PCR) in N39 cells (data not shown). teins that act as ligands for -mediated signaling Wnt3a increases insulin expression in pancreatic β-cells pathways [29]. The three well-characterized Wnt sig- and neuronal progenitor cells (derived from adult hippo- naling pathways are the canonical Wnt pathway, the campus or the olfactory bulb) via Wnt/β-catenin signaling non-canonical planar pathway, and the [35, 47]. Therefore, we tested whether Wnt3a can induce non-canonical Wnt/calcium pathway. insulin production in N39 cells. The canonical Wnt signaling regulates the expression N39 cells were treated with Wnt3a (25 or 100 ng/mL) of insulin in pancreatic β-cells and the development of for 6, 12, or 24 h. After Wnt3a treatment for 6 h, the the pancreas [30–35]. Without Wnt ligands, β-catenin, Ins2 mRNA level increased significantly (by 4.1-fold at which enhances the expression of many genes in the nu- 25 ng/mL and 3.5-fold at 100 ng/mL Wnt3a; Fig. 1a), cleus, is degraded by the β-catenin destruction complex although this increase was not dose-dependent. After [36]. Glycogen synthase kinase 3 (GSK3), a key compo- 12 h, Wnt3a increased the Ins2 mRNA level by 3.9-fold nent of the β-catenin destruction complex, phosphory- at 25 ng/mL and 4.7-fold at 100 ng/mL. After 24 h, the lates β-catenin at Ser33, Ser37, and Thr41, leading to its Ins2 mRNA level was increased by 5.4-fold at 25 ng/mL degradation by the [37]. When Wnt ligands and 6.8-fold at 100 ng/mL of Wnt3a compared to vehicle- bind to (Fz) receptor and its coreceptor, low treated cells. Thus, the Ins2 mRNA levels were significantly density lipoprotein receptor–related protein (LRP), the increased in Wnt3a-treated N39 cells in a time-dependent β-catenin destruction complex is disrupted by GSK3 manner at 12 and 24 h. inactivation [38]. Consequently, β-catenin accumulates in To confirm the expression of insulin, we checked its the cytosol and translocates into the nucleus to enhance protein level using immunofluorescence analysis with the expression of many target genes [39]. an antibody against proinsulin and found stronger The activity of the canonical Wnt signaling is also proinsulin-positive signals in N39 cells treated with observed in the adult mice hypothalamus [40, 41]. Wnt- Wnt3a (100 ng/mL) for 24 h than in vehicle-treated responsive cells are abundant in the paraventricular and cells (Fig. 1b). arcuate nuclei of the hypothalamus [42, 43]. In the mouse To further verify the production of insulin in Wnt3a- models of metabolic diseases such as diet-induced obese treated N39 cells, culture medium was collected after (DIO) mouse model and leptin-deficient (ob/ob) mouse Wnt3a treatment for 24 h, concentrated, and the concen- model, hypothalamic Wnt signaling is disrupted [44]. tration of immuno-reactive insulin was measured with GSK3 is increased in the hypothalamus of DIO mice, and ELISA. The concentration of insulin in culture medium a GSK3 inhibitor acutely improves glucose metabolism in was significantly increased (by 17.4 %) in cells treated with these mice [44]. In addition, the phosphorylated (active) Wnt3a (100 ng/mL) as compared to vehicle-treated cells form of LRP is down-regulated in ob/ob mice [45]. Down- (Fig. 1c). regulated Wnt signaling is reinstated by leptin treatment Taken together, our results indicate that Wnt3a in- [45]. These reports suggest that Wnt signaling is necessary creases the levels of insulin mRNA and protein in N39 for glucose homeostasis in the hypothalamus. However, cells. Furthermore, Wnt3a also increases of the specific mechanism how Wnt signaling modulates me- insulin from N39 cells. tabolism in the hypothalamus has not yet been revealed. Insulin signaling plays many roles in the brain [46], Wnt3a induces insulin production in N39 cells through however, the mechanisms underlying the expression of the canonical Wnt/β-catenin signaling insulin in the brain have not been discovered yet. In this In the canonical Wnt signaling, Wnt ligands bind to study, we examined how Wnt signaling regulates the Fz receptor and LRP, and this binding results in the production of insulin in the hypothalamus. The results stabilization, accumulation, and nuclear translocation of our research will help to reveal the mechanisms of β-catenin [48, 49]. To determine whether Wnt3a Lee et al. Molecular Brain (2016) 9:24 Page 3 of 12

Fig. 1 Wnt3a upregulates insulin synthesis and secretion from N39, a hypothalamic neuronal cell line. a N39 cells were treated with vehicle (PBS) or Wnt3a (25 or 100 ng/mL) for 6, 12, and 24 h, and the levels of Ins2 mRNA were measured by qRT-PCR and normalized to the levels of GAPDH mRNA (n = 23). b N39 cells were treated with vehicle or Wnt3a (100 ng/mL) for 24 h, and the induction of insulin was examined by immunofluorescence analysis with an antibody against proinsulin. Nuclei were stained with Hoechst 33342 dye. Scale bar, 20 μm. c Culture media were collected for 24 h after treatment with vehicle or Wnt3a (100 ng/mL), concentrated in Vivaspin columns, and insulin concentrations were measured by ELISA (n =6).Dataare means + SEM. *p <0.05,**p <0.01,***p <0.001comparedwithvehicle control at each time point induced Ins2 through the canonical Wnt signaling, we Consistent with the immunoblotting data, immunofluor- measured the level of β-catenin in Wnt3a-treated N39 escence analysis showed that Wnt3a treatment resulted cells. in the accumulation of active β-catenin (Fig. 2d). Fur- Immunoblot analysis (Fig. 2a-c) showed that Wnt3a thermore, active β-catenin was clearly co-localized with strongly stabilized β-catenin in a dose-dependent man- Hoechst 33342, a nuclear marker, in Wnt3a-treated N39 ner, as shown by a significant increase in the levels of cells. active β-catenin, in which the Ser33, Ser37, and Thr41 To ascertain the role of GSK3 in insulin production residues are dephosphorylated. Consequently, total β- stimulated by Wnt3a treatment in N39 cells, we treated catenin also accumulated in a dose-dependent manner. them with 6-bromoindirubin-3′-oxime (BIO), a specific Lee et al. Molecular Brain (2016) 9:24 Page 4 of 12

Fig. 2 (See legend on next page.) Lee et al. Molecular Brain (2016) 9:24 Page 5 of 12

(See figure on previous page.) Fig. 2 Wnt3a induces the expression of Ins2 by activating Wnt/β-catenin signaling. a N39 cells were treated with vehicle or Wnt3a (25 or 100 ng/mL) for 12 and 24 h. Active (Non-phospho) β-catenin and total β-catenin were detected using immunoblot assay. b and c The intensity of bands shown in (a) was quantified by using the ImageJ software with normalization to GAPDH (n =6).d N39 cells were treated with vehicle or Wnt3a (100 ng/mL) for 24 h, and active β-catenin was examined by immunofluorescence analysis; Hoechst 33342 dye was used for nuclear staining. Scale bar, 20 μm. e N39 cells were treated with 1 μM BIO, a GSK3 inhibitor, for 12 h, and the level of Ins2 mRNA was measured by qRT-PCR (n =9).f N39 cells were treated with vehicle (DMSO) or 1 μM BIO for 1, 3, 6, and 12 h to observe accumulation of active β-catenin and total β-catenin (n =6).g and h Quantification of immunoblot data in (f) using the ImageJ software was performed with normalization to GAPDH. Data are means + SEM. **p < 0.01, ***p < 0.001 compared with vehicle treatment

GSK3 inhibitor. BIO inhibits GSK3 by interacting with Induction of insulin production by Wnt3a depends on its ATP binding pocket, leading to activation of the ca- NeuroD1 nonical Wnt signaling in and mouse embryonic To unravel the mechanism of insulin production in N39 stem cells [50, 51]. BIO treatment increased the level of cells, we investigated the downstream targets of Wnt/ Ins2 mRNA by 2-fold after treatment for 12 h (Fig. 2e). β-catenin signaling. Among many targets of this path- As expected, the levels of both active and total β-catenin way, we tested NeuroD1, which is known as a candidate increased by up to 2-fold after BIO treatment for 3 h, involved in the expression of the insu- and the increased level was maintained after BIO treat- lin genes in the pancreas [52]. ment for 6 and 12 h (Fig. 2f-h). The induction of Ins2 We measured the NeuroD1 mRNA level at 6, 12, and and accumulation of β-catenin upon BIO treatment 24 h in N39 cells treated with Wnt3a (25 or 100 ng/mL). showed trends similar to those of the effects of Wnt3a At 6 h, the level of NeuroD1 mRNA was significantly treatment. Collectively, these data indicate that Wnt3a increased in comparison with vehicle-treated N39 (by increases insulin production by activating β-catenin and 3.7-fold at 25 ng/mL and 3.5-fold at 100 ng/mL Wnt3a; it might be mediated by inactivation of GSK3. Fig. 3a). At 12 h, NeuroD1 mRNA was upregulated by

Fig. 3 Wnt3a increases the expression of NeuroD1 by activating Wnt/β-catenin signaling in N39 cells. a NeuroD1 mRNA levels were quantified using qRT-PCR after vehicle or Wnt3a (25 or 100 ng/mL) treatment for 6, 12, and 24 h (n = 23). b NeuroD1 protein levels were measured by immunoblot assay after vehicle or Wnt3a (20 or 100 ng/mL) treatment for 12 and 24 h. c The intensity of bands shown in (b) was quantified by using ImageJ with normalization to GAPDH (n =6).d After treatment with vehicle or Wnt3a (100 ng/mL) for 24 h, NeuroD1 was examined using the immunofluorescence assay; the nuclei were stained with Hoechst 33342 dye. Scale bar, 20 μm. e The level of NeuroD1 mRNA was determined by qRT-PCR 12 h after treatment with 1 μMBIO(n =9).f Cells were treated with 1 μM BIO, and the level of NeuroD1 was measured by immunoblot assay after treatment for 1, 3, 6, and 12 h. g The intensity of NeuroD1 bands in (f) was quantified using the ImageJ software andnormalizedtoGAPDH(n =6).Dataaremeans+SEM.*p < 0.05, **p < 0.01, ***p < 0.001 compared with vehicle treatment Lee et al. Molecular Brain (2016) 9:24 Page 6 of 12

3.9-fold at 25 ng/mL and 5.1-fold at 100 ng/mL Wnt3a. At 24 h, the effects on NeuroD1 mRNA were even more profound (4.1-fold increase at 25 ng/mL and 8.0-fold in- crease at 100 ng/mL Wnt3a). The induction of NeuroD1 mRNA by Wnt3a showed a pattern similar to that of Ins2 mRNA: after Wnt3a treatment for 6 h, Ins2 and NeuroD1 did not show dose-dependent effects, whereas both the Ins2 and NeuroD1 mRNA levels were signifi- cantly increased in a dose-dependent manner at 12 and 24 h. Interestingly, at 1.5 h after Wnt3a (100 ng/mL) treatment, the mRNA level of NeuroD1 was increased significantly whereas that of Ins2 was not increased yet (Additional file 1: Figure S1A and S1B). To validate the effect of Wnt3a on NeuroD1 expression, we determined the level of the NeuroD1 protein in Wnt3a-treated N39 cells by immunoblotting (Fig. 3b and c). At 12 h, the NeuroD1 level was significantly increased (by 3.2 fold at 25 ng/mL and 3.8 fold at 100 ng/mL Wnt3a). At 24 h, the increase was dose-dependent (2.6 fold at 25 ng/mL and 5.4 fold at 100 ng/mL Wnt3a). NeuroD1 was detectable in N39 cells by immunofluor- escence analysis (Fig. 3d). When the cells were treated with Wnt3a (100 ng/mL) for 24 h, the intensity of the NeuroD1-positive signal was apparently greater. Co- localized with Hoechst 33342 demonstrated Wnt3a- induced nuclear localization of NeuroD1. These data demonstrate that Wnt3a increases both the expression of NeuroD1 and its translocation into the nucleus in N39 cells. We also treated N39 cells with BIO to determine whether the induction of NeuroD1 is mediated by GSK3- dependent Wnt/β-catenin signaling. The level of NeuroD1 mRNA was increased by 2-fold in cells treated with BIO for 12 h (Fig. 3e). In a similar manner, BIO treatment for 3, 6, or 12 h significantly increased the levels of the Neu- roD1 protein (by up to 2-fold compared to vehicle-treated N39 cells; Fig. 3f and g). These results suggest that Wnt/ β-catenin signaling increases the expression of NeuroD1 through GSK3 inhibition. To determine whether the induction of BDI by Wnt3a is NeuroD1-dependent, knockdown experiments were de- signed using a lentiviral vector system. Among 5 candidate short-hairpin RNA (shRNA) constructs designed to target NeuroD1, two shRNAs were chosen because of their Fig. 4 Knockdown of NeuroD1 reduces the expression of Ins2. high knockdown efficiency and used in combination for NeuroD1 was knocked down by infecting N39 cells with lentivirus NeuroD1 knockdown (data not shown). Lentiviruses containing shRNA (shNeuroD1) for 72 h. a Knockdown efficiency NeuroD1 expressing NeuroD1 shRNAs significantly suppressed was determined by examining the levels of mRNA using qRT-PCR (n = 23). b The level of NeuroD1 protein was assessed by the expression of NeuroD1: by 80 % at the mRNA level immunoblot assay, and c the intensity of NeuroD1 bands was (Fig. 4a) and by 50 % at the protein level after 72 h quantified using the ImageJ software with normalization to GAPDH infection (Fig. 4b and c) compared to lentivirus ex- (n =6).d The basal expression of Ins2 was determined by qRT-PCR pressing non-targeting shRNA. Interestingly, NeuroD1 after infection with shRNA lentivirus against NeuroD1 for 72 h n p p knockdown reduced the expression of Ins2 by 40 % of ( = 12). Data are means + SEM. ** < 0.01, *** < 0.001 compared with non-targeting shRNA lentivirus-infected cells (shNon-target) that in non-targeting shRNA lentivirus–infected cells in the absence of Wnt3a, indicating that NeuroD1 is Lee et al. Molecular Brain (2016) 9:24 Page 7 of 12

involved in maintaining the expression of Ins2 in the basal state (Fig. 4d). In addition, we treated N39 cells, in which NeuroD1 was knocked down, with Wnt3a (100 ng/mL) to examine any changes in the induction of Ins2. As shown in Fig. 5a and b, Wnt3a treatment resulted in the accumulation of active and total β-catenin in cells infected with lentiviruses expressing either non-targeting shRNA or NeuroD1shRNA.However,asdemonstratedbyusing qRT-PCR, the NeuroD1 transcript was not upregulated by Wnt3a in N39 cells with knocked down NeuroD1, although it was significantly increased in non-targeting shRNA lentivirus–infected N39 cells (Fig. 5c). Further- more, Wnt3a treatment failed to induce Ins2 in N39 cells with knocked down NeuroD1 whereas Wnt3a significantly increased the level of Ins2 mRNA in non-targeting shRNA lentivirus–infected N39 cells (Fig. 5d). Taken together, these results suggest that NeuroD1 is a key transcription factor for Wnt3a-induced Ins2 expres- sion through Wnt/β-catenin signaling in N39 cells.

Wnt3a administration induces the expression of Ins2 and NeuroD1 in the hypothalamus Many reports showed Wnt activity in the adult hypothal- amus [40–43]. Autocrine and/or paracrine Wnt ligands may be responsible for Wnt activity in the hypothalamus. To test Wnt3a-induced BDI production in the hypothal- amus in vivo, we first measured the mRNA level of Wnt3a by qRT-PCR in the hypothalamic tissues (Fig. 6a). Al- though the amount of Wnt3a mRNA was relatively lower than that of Ins2, the expression of Wnt3a was detected in the hypothalamus. To examine the effect of Wnt3a treatment on the syn- thesis of Ins2 mRNA in hypothalamic tissue, we introduced Wnt3a into the hypothalamus by intracerebroventricular (icv) injection and collected hypothalamic tissues for ana- lysis 24 h after injection. Administration of Wnt3a (4 or 20 ng) significantly increased the levels of Ins2 mRNA (by 1.3-fold and 1.6-fold, respectively; Fig. 6b). Interest- Fig. 5 Wnt3a-induced Ins2 upregulation is blocked by knockdown of ingly, the levels of NeuroD1 mRNA were also upregulated NeuroD1. NeuroD1 was knocked down by infecting N39 cells with (Fig. 6c). These results indicate that Wnt3a increases the lentivirus containing shRNA for 72 h. a N39 cells were infected by lentiviral shRNA against NeuroD1 for 48 h and then treated with expression of Ins2 and NeuroD1 in the hypothalamus. Wnt3a (100 ng/mL) for 24 h. The levels of active and total β-catenin and NeuroD1 protein were determined by immunoblot assay. b The Discussion intensity of NeuroD1 signal in (a) was quantified using the ImageJ Despite accumulating evidence about BDI, the specific software (n = 4). The levels of NeuroD1 mRNA (c) and Ins2 mRNA (d) mechanism of BDI production is not yet well under- were measured by qRT-PCR in N39 cells that were infected with lentiviral non-targeting or NeuroD1 shRNA for 48 h and then treated stood. In particular, hypothalamic insulin production has with Wnt3a (100 ng/mL) for 24 h (n = 9). Data are means + SEM. not been reported. Considering that the hypothalamus ***p < 0.001 compared with vehicle treatment in shNon-target. has a large number of neurosecretory cells, which produce ###p < 0.001 compared with Wnt3a treatment in shNon-target. and secrete diverse involved in maintaining N.S., not significant within shNeuroD1 samples homeostasis, it is plausible that BDI plays a specific role in maintaining energy homeostasis and regulating me- hypothalamus by activating Wnt/β-catenin signaling and tabolism in the hypothalamus. In this study, we demon- NeuroD1 induction. Inhibition of GSK3, which phosphor- strated that Wnt3a increases insulin production in the ylates β-catenin and thus promotes its degradation, leads Lee et al. Molecular Brain (2016) 9:24 Page 8 of 12

Synthesized NeuroD1 also moves into the nucleus and induces insulin production (Fig. 7). Hippocampus and olfactory regions are known as brain regions that express insulin higher than other brain re- gions [15, 35]. The Ins2 mRNA level is lower in the hypo- than in the hippocampus [14, 17]. However, insulin signaling plays important roles in controlling food intake, glucose metabolism, and energy expenditure in the hypothalamus [53, 54]. Nevertheless the expression level is lower than other regions, we hypothesized that BDI syn- thesized in the hypothalamus could regulate metabolism by acting on the hypothalamic neurocircuits. The activation of Wnt/β-catenin signaling enhances the transcription of downstream genes via the formation of a complex between active β-catenin and TCF [55]. Our results demonstrate not only the accumulation of active and total β- catenin (Fig. 2a-c), but also the translocation of active β- catenin into the nucleus induced by Wnt3a in N39 cells (Fig. 2d). Although we did not examine the activity of the TCF complex, these results indicate activation of Wnt/β- catenin signaling in these cells. The level of NeuroD1, a downstream target of Wnt/β-catenin signaling, was in- creased in Wnt3a-treated N39 cells (Fig. 3a-d). This in- crease in NeuroD1 expression supports that Wnt3a- induced accumulation of active β-catenin enhances the transcription of NeuroD1 gene. To determine whether GSK3 is involved in Wnt3a- induced Ins2 upregulation, we treated BIO to inhibit GSK3 (Fig. 2e-h, Fig. 3e-g). BIO treatment resulted in accumulation of β-catenin, and increases in the ex- pression of NeuroD1 and Ins2, suggesting that GSK3 plays a role in the upregulation of Ins2 by Wnt/β-ca- tenin signaling. The of GSK3 at Ser9 residue is one of mechanisms for GSK3 inactivation [56, 57]. However, we could not observe any changes in the level of phospho-GSK3 (Ser9) when N39 was treated with Wnt3a (data not shown). Therefore, it is likely that GSK3 inhibition by Wnt3a is mediated by not phosphorylation of Ser9 but other mechanisms in our study. NeuroD1 is a basic helix-loop-helix transcription factor that plays a role in facilitating differentiation of neurons Fig. 6 Wnt3a upregulates NeuroD1 and Ins2 in the mouse [58, 59], and its gene is a target of Wnt/β-catenin signaling hypothalamus. The basal mRNA levels of Ins2 and Wnt3a were in the brain [52, 60]. NeuroD1 also binds to the insulin pro- measured by qRT-PCR in hypothalamic tissues from C57BL/6 male moter to increase the expression [61, 62]. Although an in- n − mice ( =6 9) (a). C57BL/6 male mice received icv injections of ve- crease in Ins2 mRNA levels in Wnt3a-injected hypothalamic hicle or Wnt3a (4 or 20 ng), and hypothalamic tissues were dissected after 24 h. The mRNA levels of Ins2 (b) and NeuroD1 (c) were mea- tissues was smaller than that in N39 cells, the levels of both sured by qRT-PCR (n =6−9). Data are means + SEM. ###p < 0.001 NeuroD1 and Ins2 mRNAs increased in the hypothalamus of compared with Ins2 mRNA, ***p < 0.001 compared with Wnt3a-injected mice. Taken together, it is likely that insulin vehicle administration production induced by Wnt3a, which is well known to occur in the pancreas, is conserved in the hypothalamus. to accumulation of β-catenin in the cytosol. Accumulated The mechanism of insulin secretion in neurons is not β-catenin translocates into the nucleus, where it enhances yet revealed. Mechanistic studies on insulin secretion in the expression of NeuroD1 with T-cell factor (TCF). vivo are challenging, because it is difficult to pinpoint Lee et al. Molecular Brain (2016) 9:24 Page 9 of 12

Fig. 7 The proposed regulatory mechanism of insulin expression by the Wnt/β-catenin/NeuroD1 pathway in the hypothalamus. Wnt3a might bind to LRP and Frizzed receptor to activate the canonical Wnt pathway. GSK3 phosphorylates β-catenin and induces its degradation; Wnt3a inhibits GSK3, leading to β-catenin accumulation. Accumulated β-catenin translocates into the nucleus and enhances the expression of NeuroD1 with TCF. Up-regulated NeuroD1 also translocates into the nucleus, where it induces the production of insulin the origin of insulin in the brain; the confounding factor Conclusions is that insulin produced in the pancreas is delivered by We demonstrate that BDI is regulated by Wnt/β-catenin blood circulation to the brain in vivo. Furthermore, the signaling in the hypothalamus. It is possible that insulin basal expression of insulin is not high enough to study production in the hypothalamus or other brain areas the mechanism of its secretion in in vitro systems. changes in metabolic diseases. Further studies focused on Insulin-overexpressing N39 cells as an in vitro model the regulation of BDI production in the disease models might be useful to investigate the mechanism of insulin with disrupted hypothalamic Wnt/β-catenin signaling will secretion in the future studies. be needed to elucidate the pathophysiological roles of The neuron-specific insulin receptor knockout (NIRKO) BDI. mice resulted in increased food intake and moderate diet- induced , demonstrating the anti-obesity role of Methods insulin [2]. However, insulin receptor (IR) knockout Cell culture and treatments mice specifically deleted in agouti-related peptide (AgRP)- Immortalized mouse hypothalamic cell line N39 (mHypoE- or pro-opiomelanocortin (POMC)-expressing neuron did 39) was obtained from CELLutions Biosystems and main- not show a significant difference in energy homeostasis, tained in Dulbecco’s modified Eagle medium (DMEM) especially food intake [63]. Moreover, deletion of IR in (Sigma) supplemented with 10 % fetal bovine serum (FBS) steroidogenic factor 1 (SF1)-expressing neurons of the (Hyclone Laboratories) and 1 % penicillin/streptomycin ventromedial hypothalamus had no effect on body weight (GIBCO). Lenti-X™ 293 T cell line was purchased from and food intake [64]. The different phenotypes among Clontech and cultured in DMEM supplemented with mice models suggest that the roles of insulin through IR 10 % FBS and 1 % penicillin/streptomycin. Recombinant in the brain are diverse in a cell-type specific manner. mouse Wnt3a (1324-WN, R&D Systems) was dissolved in Therefore, it might be important to identify the subpopu- -buffered saline (PBS) containing 0.2 % BSA. lation of hypothalamic neurons that specifically express BIO (Sigma), a specific GSK3 inhibitor, was dissolved in BDI in order to investigate the role of BDI in the future dimethyl sulfoxide (DMSO) at 1 mM concentration. study. It would be interesting to investigate whether Wnt3a-induced insulin production from the specific neu- rons in the hypothalamus can regulate the food intake and studies were performed in accordance with the protect from obesity. guidelines on care and use as approved by the DGIST Lee et al. Molecular Brain (2016) 9:24 Page 10 of 12

Institutional Animal Care and Use Committee. C57BL/6 sodium pyrophosphate, 1 mM EDTA, 1 mM EGTA, 1 % mice were obtained from Koatech; 8-week-old males Triton X-100, 0.1 mM benzamidine, 10 μg/mL leupeptin, were used for experiments. Mice were housed in groups 1 mM DTT, 0.5 mM PMSF, 50 mM NaF, 1X protease of 3–5 under a 12/12 h light/dark cycle (lights on from inhibitor cocktail (Calbiochem), 1X phosphatase inhibitor 6:00 to 18:00) in an individually ventilated cage Innorack cocktail (Sigma). Protein was extracted from hypothalamic (Innovive). tissues with the same lysis buffer. Lysates (20 μg total pro- tein) were separated on 10 % SDS-polyacrylamide gels and Administration of Wnt3a and preparation of brain tissues blotted onto polyvinylidene difluoride membranes for Wnt3a was introduced into the hypothalamus by icv 40 min at 20 V in transfer buffer containing 25 mM Tris injection of 2 μL of PBS containing Wnt3a (4 or 20 ng) base and 192 mM glycine. The membranes were blocked into the third ventricles. Mice were euthanized with CO2 with 5 % skim milk for 3 h and then incubated with gas supplied for 3–5 min 24 h after Wnt3a administra- primary antibodies against non-phospho β-catenin (1:3000; tion. Brain was quickly removed, placed on ice, and the ), total β-catenin (1:3000; Cell Signaling), hypothalamus was dissected using as landmarks the optic NeuroD1 (1:3000; AbCam), or glyceraldehyde-3-phosphate chiasm and the mammillary bodies to depth of ~ 2 mm dehydrogenase (GAPDH) (1:5000; Cell Signaling) at 4 °C [65]. Hypothalamic tissues were frozen in liquid nitrogen overnight. After three washes with Tris-buffered saline and ground mechanically on dry ice. with 0.1 % of Tween 20 (TBST), the membranes were incubated with horseradish peroxidase–linked second- RNA extraction and gene expression analysis by ary antibody and washed with TBST three times and quantitative real-time PCR visualized by SuperSignal West Pico Chemiluminescent Total RNA was isolated using TRIzol (Invitrogen) and Substrate (Thermo) according to the manufacturer’s chloroform (Sigma) from N39 cells and ground brain procedure. The intensity of bands was quantified using tissues. For qRT-PCR analysis, 3 μg of total RNA from NIH ImageJ analysis software. each sample was reverse-transcribed using a Reverse Transcription System (Promega). Synthesized cDNA was Immunofluorescence analysis diluted 1:5 with water, and 2 μl was used as a tem- N39 cells were seeded on poly-L-lysine–coated coverslips. plate for qRT-PCR. Primers were designed as follows: After treatment with Wnt3a for 24 h, cells were washed Ins1 Forward, 5′-AGAGACCATCAGCAAGCAGGTCA- with PBS, fixed with 4 % paraformaldehyde and perme- 3′; Ins1 Reverse, 5′-TACCAGGTGGGGACCACAAA abilized in 0.2 % Triton X-100, 0.1 M glycine. The cells GA-3′; Ins2 Forward, 5′-GTGACCTTCAGACCTTGG were then washed with PBS, blocked with BSA containing CACTG-3′; Ins2 Reverse, 5′-AGGCTGGGTAGTGGT antibody diluent (Invitrogen) at room temperature (RT) GGGTCTAG-3′; NeuroD1 Forward, 5′-TGACCTTTC and incubated with antibodies against proinsulin (1:50; CCATGCTGAAT-3′; NeuroD1 Reverse, 5′-AAGTGC R&D Systems), non-phospho β-catenin (1:500; Cell Signal- TAAGGCAACGCAAT-3′; GAPDH Forward, 5′-GTC ing), and NeuroD1 (1:200; Santa Cruz) at 4 °C. Coverslips AATGAAGGGGTCGTTGATGG-3′;andGAPDH Reverse, were incubated with secondary antibodies for 2 h at RT, 5′-TCGTCCCGTAGACAAAATGGTGA-3′.qRT-PCRwas and Hoechst 33342 (Invitrogen) was used for nuclear performed according to the SYBR Green protocol (SYBR staining for 10 min at RT. ProLong Diamond Antifade Premix Ex Taq, TaKaRa) in a Bio-Rad CFX-96 machine. An- Mountant (Invitrogen) was used for mounting coverslips. nealing and extension were done at 63 °C. Each slide was analyzed with an LSM700 confocal micro- scope (Carl Zeiss). Images were analyzed with ZEN2012 Enzyme-linked immunosorbent assay software (Carl Zeiss). Culture medium from Wnt3a-treated N39 cells was col- lected and filtered with a 0.22 μm syringe filter. Vivaspin 2 Preperation of lentiviral shRNAs (2000 MWCO Hydrosart; Sartorius) was used to concen- Five TRC Lentiviral Non-targeting shRNAs control trate 3 mL of medium to 35 μL. Insulin concentrations in (#RHS6848) and mouse NeuroD1 shRNAs (81773-81777, concentrated media were measured by using an ELISA Dharmacon) were cloned into the pLKO.1 lentiviral vector. , which detects only mature insulin, purchased from ThepMD2.GandpsPAX2vectorswereusedasanenve- ALPCO Diagnostics. lope and packaging vector, respectively, to produce lenti- virus in the Lenti-X 293 T cell line (Clontech). Transient Immunoblot analysis transfections were performed with TurboFect Transfection N39 cells were plated at a density of 1.0 × 105/mL in Reagent (Thermo) according to the manufacturer’sproto- 6-well plates. After appropriate treatments, cells were col. Cell supernatants containing virus were collected 72 h washed with ice-cold PBS and harvested in lysis buffer post-transfection and filtered through 0.45 μmsyringe (50 mM Tris-HCl, pH 7.4, 250 mM sucrose, 5 mM filters. Lentiviral vectors were concentrated by Lee et al. Molecular Brain (2016) 9:24 Page 11 of 12

ultracentrifugation at 40,000 × g for 90 min at 4 °C. Viral 9. Moloney AM, Griffin RJ, Timmons S, O'Connor R, Ravid R, O'Neill C. Defects pellets were suspended in PBS. in IGF-1 receptor, insulin receptor and IRS-1/2 in Alzheimer’s disease indicate possible resistance to IGF-1 and insulin signalling. Neurobiol Aging. 2010;31(2):224–43. doi:10.1016/j.neurobiolaging.2008.04.002. Statistical analysis 10. Liu Y, Liu F, Grundke-Iqbal I, Iqbal K, Gong CX. Deficient brain insulin signalling pathway in Alzheimer’s disease and . J Pathol. 2011;225(1):54–62. Data were analyzed with GraphPad Prism software (Graph- doi:10.1002/path.2912. Pad Software). Statistical analysis was performed using 11. Margolis RU, Altszuler N. Insulin in the cerebrospinal fluid. Nature. unpaired t-test. 1967;215(5108):1375–6. 12. Banks WA, Jaspan JB, Huang W, Kastin AJ. Transport of insulin across the blood-brain barrier: saturability at euglycemic doses of insulin. Peptides. Additional file 1997;18(9):1423–9. 13. Havrankova J, Roth J, Brownstein M. Insulin receptors are widely distributed in the of the rat. Nature. 1978;272(5656):827–9. Additional file 1: Supplemental Figure S1. Wnt3a induces NeoruD1 14. Madadi G, Dalvi PS, Belsham DD. Regulation of brain insulin mRNA by prior to upregulation of Ins2. N39 cells were treated with vehicle (PBS) glucose and glucagon-like peptide 1. Biochem Biophys Res Commun. or Wnt3a (25 or 100 ng/mL) for 1.5, 3 and 4.5h, and the levels of NeoruD1 2008;376(4):694–9. doi:10.1016/j.bbrc.2008.09.054. mRNA (A) and Ins2 mRNA (B) were measured by qRT-PCR and normalized 15. Mehran AE, Templeman NM, Brigidi GS, Lim GE, Chu KY, Hu X, et al. to the levels of GAPDH mRNA (n=9). Data are means + SEM. *p < 0.05, ** Hyperinsulinemia drives diet-induced obesity independently of brain insulin p < 0.01, *** p < 0.001 compared with vehicle control at each time point. production. Cell Metab. 2012;16(6):723–37. doi:10.1016/j.cmet.2012.10.019. (JPG 976 kb) 16. Shiao MS, Liao BY, Long M, Yu HT. Adaptive evolution of the insulin two-gene system in mouse. Genetics. 2008;178(3):1683–91. doi:10.1534/genetics.108.087023. Abbreviations 17. Young 3rd WS. Periventricular hypothalamic cells in the rat brain contain – AgRP: agouti-related peptide; BDI: brain-derived insulin; DIO: diet-induced insulin mRNA. Neuropeptides. 1986;8(2):93 7. obese; Fz: frizzled; GSK3: glycogen synthase kinase 3; IR: insulin receptor; 18. Devaskar SU, Singh BS, Carnaghi LR, Rajakumar PA, Giddings SJ. Insulin II gene – LRP: low density lipoprotein receptor-related protein; N39 cell: mouse expression in rat central nervous system. Regul Pept. 1993;48(1-2):55 63. hypothalamic neuronal cell; NeuroD1: neurogenic differentiation 1; 19. Devaskar SU, Giddings SJ, Rajakumar PA, Carnaghi LR, Menon RK, Zahm DS. NIRKO: neuron-specific insulin receptor knockout; ob/ob: leptin-deficient; Insulin gene expression and insulin synthesis in mammalian neuronal cells. – POMC: pro-opiomelanocortin; SF1: steroidogenic factor 1; TCF: T-cell factor. J Biol Chem. 1994;269(11):8445 54. 20. Singh BS, Rajakumar PA, Eves EM, Rosner MR, Wainer BH, Devaskar SU. Insulin gene expression in immortalized rat hippocampal and -12 Competing interests cell lines. Regul Pept. 1997;69(1):7–14. The authors declare that they have no conflict of interests. 21. Hrytsenko O, Wright Jr JR, Morrison CM, Pohajdak B. Insulin expression in the brain and pituitary cells of tilapia (Oreochromis niloticus). Brain Res. Authors’ contributions 2007;1135(1):31–40. doi:10.1016/j.brainres.2006.12.009. JL and EKK conceived and designed the study, interpreted the results, and 22. Gerozissis K. Brain insulin, energy and glucose homeostasis; genes, wrote the manuscript. JL and KK planned and performed experiments and environment and metabolic pathologies. Eur J Pharmacol. 2008;585(1):38–49. analyzed the results. SWY discussed the study and wrote the manuscript. doi:10.1016/j.ejphar.2008.01.050. All authors read and approved the final manuscript. 23. Molnar G, Farago N, Kocsis AK, Rozsa M, Lovas S, Boldog E, et al. GABAergic neurogliaform cells represent local sources of insulin in the cerebral cortex. Acknowledgements J Neurosci. 2014;34(4):1133–7. doi:10.1523/JNEUROSCI.4082-13.2014. This work was supported by the National Research Foundation of South 24. Dorn A, Rinne A, Bernstein HG, Hahn HJ, Ziegler M. Insulin and C-peptide in Korea (Grant No. 2013M3C7A1056099 and 2012M3A9C6049935) and the human brain neurons (insulin/C-peptide/brain peptides/ DGIST R&D Program of the Ministry of Science, Information and Communication /radioimmunoassay). J Hirnforsch. 1983;24(5):495–9. Technology, and Future Planning of South Korea (Grant No. 15-BD-0402). 25. Frolich L, Blum-Degen D, Bernstein HG, Engelsberger S, Humrich J, Laufer S, et al. Brain insulin and insulin receptors in aging and sporadic Alzheimer’s Received: 10 December 2015 Accepted: 26 February 2016 disease. J Neural Transm. 1998;105(4-5):423–38. 26. Lee SM, Tole S, Grove E, McMahon AP. A local Wnt-3a signal is required for development of the mammalian hippocampus. References Development. 2000;127(3):457–67. 1. Schechter R, Whitmire J, Holtzclaw L, George M, Harlow R, Devaskar SU. 27. Galceran J, Miyashita-Lin EM, Devaney E, Rubenstein JL, Grosschedl R. Developmental regulation of insulin in the mammalian central nervous system. Hippocampus development and generation of dentate gyrus cells Brain Res. 1992;582(1):27–37. is regulated by LEF1. Development. 2000;127(3):469–82. 2. Bruning JC, Gautam D, Burks DJ, Gillette J, Schubert M, Orban PC, et al. Role 28. Zhou CJ, Zhao C, Pleasure SJ. Wnt signaling mutants have decreased of brain insulin receptor in control of body weight and reproduction. dentate granule cell production and radial glial scaffolding abnormalities. Science. 2000;289(5487):2122–5. J Neurosci. 2004;24(1):121–6. doi:10.1523/JNEUROSCI.4071-03.2004. 3. Woods SC, Seeley RJ, Baskin DG, Schwartz MW. Insulin and the blood-brain 29. Logan CY, Nusse R. The Wnt signaling pathway in development and barrier. Curr Pharm Des. 2003;9(10):795–800. disease. Annu Rev Cell Dev Biol. 2004;20:781–810. doi:10.1146/annurev. 4. Fisher SJ, Bruning JC, Lannon S, Kahn CR. Insulin signaling in the central cellbio.20.010403.113126. nervous system is critical for the normal sympathoadrenal response to 30. Sander M, German MS. The beta cell transcription factors and development . Diabetes. 2005;54(5):1447–51. of the pancreas. J Mol Med (Berl). 1997;75(5):327–40. 5. Ott ASR, van Harskamp F, Pols HA, Hofman A, Breteler MM. Diabetes mellitus 31. Heller RS, Klein T, Ling Z, Heimberg H, Katoh M, Madsen OD, et al. and the risk of dementia: the Rotterdam study. Neurology. 1999;53(9):1937–42. Expression of Wnt, Frizzled, sFRP, and DKK genes in adult human pancreas. 6. Steen E, Terry BM, Rivera EJ, Cannon JL, Neely TR, Tavares R, et al. Impaired Gene Expr. 2003;11(3-4):141–7. insulin and insulin-like expression and signaling mechanisms in 32. Fujino T, Asaba H, Kang MJ, Ikeda Y, Sone H, Takada S, et al. Low-density Alzheimer’s disease–is this type 3 diabetes? J Alzheimers Dis. 2005;7(1):63–80. lipoprotein receptor-related protein 5 (LRP5) is essential for normal 7. Schubert M, Gautam D, Surjo D, Ueki K, Baudler S, Schubert D, et al. Role for cholesterol metabolism and glucose-induced insulin secretion. Proc Natl neuronal in neurodegenerative diseases. Proc Natl Acad Acad Sci U S A. 2003;100(1):229–34. doi:10.1073/pnas.0133792100. Sci U S A. 2004;101(9):3100–5. doi:10.1073/pnas.0308724101. 33. Papadopoulou S, Edlund H. Attenuated Wnt signaling perturbs 8. Cohen E, Dillin A. The insulin paradox: aging, proteotoxicity and pancreatic growth but not pancreatic function. Diabetes. neurodegeneration. Nat Rev Neurosci. 2008;9(10):759–67. doi:10.1038/nrn2474. 2005;54(10):2844–51. Lee et al. Molecular Brain (2016) 9:24 Page 12 of 12

34. Wells JM, Esni F, Boivin GP, Aronow BJ, Stuart W, Combs C, et al. Wnt/β- 55. Kuwabara T, Hsieh J, Muotri A, Yeo G, Warashina M, Lie DC, et al. catenin signaling is required for development of the exocrine pancreas. Wnt-mediated activation of NeuroD1 and retro-elements during adult BMC Dev Biol. 2007;7(1):4. neurogenesis. Nat Neurosci. 2009;12(9):1097–105. doi:10.1038/nn.2360. 35. Kuwabara T, Kagalwala MN, Onuma Y, Ito Y, Warashina M, Terashima K, et al. 56. Cross DA, Alessi DR, Cohen P, Andjelkovich M, Hemmings BA. Inhibition of Insulin biosynthesis in neuronal progenitors derived from adult glycogen synthase kinase-3 by insulin mediated by . Nature. hippocampus and the olfactory bulb. EMBO Mol Med. 2011;3(12):742–54. 1995;378(6559):785–9. doi:10.1038/378785a0. doi:10.1002/emmm.201100177. 57. Srivastava AK, Pandey SK. Potential mechanism(s) involved in the regulation 36. Schinner S, Ulgen F, Papewalis C, Schott M, Woelk A, Vidal-Puig A, et al. of glycogen synthesis by insulin. Mol Cell Biochem. 1998;182(1-2):135–41. Regulation of insulin secretion, gene transcription and beta cell 58. Lee JE. Basic helix-loop-helix genes in neural development. Curr Opin proliferation by adipocyte-derived Wnt signalling molecules. Diabetologia. Neurobiol. 1997;7(1):13–20. 2008;51(1):147–54. doi:10.1007/s00125-007-0848-0. 59. Crews ST, Fan CM. Remembrance of things PAS: regulation of development 37. Patel S, Doble B, Woodgett JR. Glycogen synthase kinase-3 in insulin and Wnt by bHLH-PAS proteins. Curr Opin Genet Dev. 1999;9(5):580–7. signalling: a double-edged sword? Biochem Soc Trans. 2004;32(Pt 5):803–8. 60. Mutoh H, Fung BP, Naya FJ, Tsai MJ, Nishitani J, Leiter AB. The basic helix- doi:10.1042/BST0320803. loop-helix transcription factor BETA2/NeuroD is expressed in mammalian 38. Welters HJ, Kulkarni RN. Wnt signaling: relevance to beta-cell enteroendocrine cells and activates secretin gene expression. Proc Natl biology and diabetes. Trends Endocrinol Metab. 2008;19(10):349–55. Acad Sci U S A. 1997;94(8):3560–4. doi:10.1016/j.tem.2008.08.004. 61. Khoo S, Griffen SC, Xia Y, Baer RJ, German MS, Cobb MH. Regulation of 39. Rulifson IC, Karnik SK, Heiser PW, ten Berge D, Chen H, Gu X, et al. Wnt insulin gene transcription by ERK1 and ERK2 in pancreatic beta cells. J Biol signaling regulates pancreatic beta . Proc Natl Acad Sci Chem. 2003;278(35):32969–77. doi:10.1074/jbc.M301198200. U S A. 2007;104(15):6247–52. doi:10.1073/pnas.0701509104. 62. Petersen HV, Jensen JN, Stein R, Serup P. Glucose induced MAPK signalling 40. Maretto S, Cordenonsi M, Dupont S, Braghetta P, Broccoli V, Hassan AB, influences NeuroD1-mediated activation and nuclear localization. FEBS Lett. et al. Mapping Wnt/beta-catenin signaling during mouse development 2002;528(1-3):241–5. and in colorectal tumors. Proc Natl Acad Sci U S A. 2003;100(6):3299–304. 63. Konner AC, Janoschek R, Plum L, Jordan SD, Rother E, Ma X, et al. Insulin action doi:10.1073/pnas.0434590100. in AgRP-expressing neurons is required for suppression of hepatic glucose 41. Wang X, Kopinke D, Lin J, McPherson AD, Duncan RN, Otsuna H, et al. Wnt production. Cell Metab. 2007;5(6):438–49. doi:10.1016/j.cmet.2007.05.004. signaling regulates postembryonic hypothalamic progenitor differentiation. 64. Klockener T, Hess S, Belgardt BF, Paeger L, Verhagen LA, Husch A, et al. Dev Cell. 2012;23(3):624–36. doi:10.1016/j.devcel.2012.07.012. High-fat feeding promotes obesity via insulin receptor/PI3K-dependent – 42. Kokoeva MV, Yin H, Flier JS. Neurogenesis in the hypothalamus of adult inhibition of SF-1 VMH neurons. Nat Neurosci. 2011;14(7):911 8. mice: potential role in energy balance. Science. 2005;310(5748):679–83. doi:10.1038/nn.2847. doi:10.1126/science.1115360. 65. Kim EK, Miller I, Aja S, Landree LE, Pinn M, McFadden J, et al. C75, a fatty acid 43. Pierce AA, Xu AW. De novo neurogenesis in adult hypothalamus as synthase inhibitor, reduces food intake via hypothalamic AMP-activated – a compensatory mechanism to regulate energy balance. J Neurosci. protein kinase. J Biol Chem. 2004;279(19):19970 6. doi:10.1074/jbc.M402165200. 2010;30(2):723–30. doi:10.1523/JNEUROSCI.2479-09.2010. 44. Benzler J, Ganjam GK, Kruger M, Pinkenburg O, Kutschke M, Stohr S, et al. Hypothalamic glycogen synthase kinase 3beta has a central role in the regulation of food intake and glucose metabolism. Biochem J. 2012;447(1):175–84. doi:10.1042/BJ20120834. 45. Benzler J, Andrews ZB, Pracht C, Stohr S, Shepherd PR, Grattan DR, et al. Hypothalamic WNT signalling is impaired during obesity and reinstated by leptin treatment in male mice. . 2013;154(12):4737–45. doi:10.1210/en.2013-1746. 46. Fernandez AM, Torres-Aleman I. The many faces of insulin-like peptide signalling in the brain. Nat Rev Neurosci. 2012;13(4):225–39. doi:10.1038/nrn3209. 47. Gui S, Yuan G, Wang L, Zhou L, Xue Y, Yu Y, et al. Wnt3a regulates proliferation, and function of pancreatic NIT-1 beta cells via activation of IRS2/PI3K signaling. J Cell Biochem. 2013;114(7):1488–97. doi:10.1002/jcb.24490. 48. Staal FJ, Noort Mv M, Strous GJ, Clevers HC. Wnt signals are transmitted through N-terminally dephosphorylated beta-catenin. EMBO Rep. 2002;3(1):63–8. doi:10.1093/embo-reports/kvf002. 49. Li VS, Ng SS, Boersema PJ, Low TY, Karthaus WR, Gerlach JP, et al. Wnt signaling through inhibition of beta-catenin degradation in an intact Axin1 complex. Cell. 2012;149(6):1245–56. doi:10.1016/j.cell.2012.05.002. 50. Meijer L, Skaltsounis AL, Magiatis P, Polychronopoulos P, Knockaert M, Leost M, et al. GSK-3-selective inhibitors derived from Tyrian purple indirubins. Chem Biol. 2003;10(12):1255–66. 51. Sato N, Meijer L, Skaltsounis L, Greengard P, Brivanlou AH. Maintenance of pluripotency in human and mouse embryonic stem cells through activation Submit your next manuscript to BioMed Central of Wnt signaling by a pharmacological GSK-3-specific inhibitor. Nat Med. and we will help you at every step: 2004;10(1):55–63. doi:10.1038/nm979. 52. Sharma A, Moore M, Marcora E, Lee JE, Qiu Y, Samaras S, et al. The • We accept pre-submission inquiries NeuroD1/BETA2 sequences essential for insulin gene transcription colocalize • Our selector tool helps you to find the most relevant journal with those necessary for neurogenesis and p300/CREB binding protein • binding. Mol Cell Biol. 1999;19(1):704–13. We provide round the clock customer support 53. Obici S, Zhang BB, Karkanias G, Rossetti L. Hypothalamic insulin signaling is • Convenient online submission required for inhibition of glucose production. Nat Med. 2002;8(12):1376–82. • Thorough peer review doi:10.1038/nm798. • Inclusion in PubMed and all major indexing services 54. Porte Jr D, Baskin DG, Schwartz MW. Insulin signaling in the central nervous system: a critical role in metabolic homeostasis and disease from C. elegans • Maximum visibility for your research to . Diabetes. 2005;54(5):1264–76. Submit your manuscript at www.biomedcentral.com/submit